The ability to detect biological molecules associated with enzyme activity has application in fields such as testing for biological contamination of water and food products. Of particular interest is the ability to detect biological (e.g., bacterial) contamination of water. Usually, methods for detection of bacteria such as Escherichia coli (E. coli or EC) and total coliform (TC) are based on detection of indicator enzyme activity in a broth designed to promote growth of the target organism. Accepted indicator enzymes are β-glucuronidase (β-glu) and β-galactosidase (β-gal) for EC and TC, respectively. Methods which use these enzymes rely on a reaction of the enzyme with a chromogenic or fluorogenic compound to measure the enzyme activity (for reviews, see refs. 1 and 2). In the case of β-glu or β-gal, usually a glucuronide or galactoside conjugate of a dye compound is added to the sample broth as a substrate, and if the target enzymes are present, the conjugate is converted to a free dye molecule. The enzyme-dependent conversion is detected by a change in colour or fluorescence of the free dye molecule compared to the conjugate. Some tests use soluble products detected in solution, with the coliform cells usually also suspended in solution. Others use coliform cells on the surface of a filter, membrane, or gel, usually with an insoluble dye product which adsorbs onto the support to form a coloured or fluorescent spot around colonies of target organisms (3). Some supported formats use multiple dye substrates which produce a variety of colours depending on which organisms are present.
However, the above approaches are vulnerable to sources of error, such as suitability of broth and incubation conditions for all target coliform types, as well as presence of non-target organisms which may contribute to the indicator enzyme activity. Nonetheless, the reliability of established methods is high enough that there is broad regulatory acceptance of these methods for assessment of samples ranging from meat products to drinking water.
Further, in routine or commercial uses of such substrates, detection is usually done visually by human eye, which presents significant limitations in performance. A large number of coliform cells must be present before enough substrate will be converted for the product to be visible. This requires significant incubation and growth for detection of a small initial number of cells, and a standard 100 mL sample is incubated for 24 h to provide a detection limit of one coliform cell in the initial sample. In some cases, more rapid detection is reported, but normally only with a higher detection limited accepted (e.g., 100 to 300 cells in a 100 mL sample (4,5,6)). Also, visual detection is not quantitative, and these tests are normally used in a “presence/absence” mode where the actual number of coliform cells is not determined. An exception to the latter is some plating methods, where the number of colonies is counted and therefore the number of cells in the sample quantitatively determined (3). This, however, is a very labour-intensive, time-consuming process which also requires long incubation, and has limited dynamic range.